EPJ manuscript No.

(will be inserted by the editor)

Quantum computer with cold ions in the Aubry pinned phase

D.L.Shepelyansky
Laboratoire de Physique Théorique, IRSAMC, Université de Toulouse, CNRS, UPS, 31062 Toulouse, France

Dated: 26 February 2019

arXiv:1902.09850v1 [quant-ph] 26 Feb 2019

Abstract. It is proposed to modify the Cirac-Zoller proposal of quantum computer with cold ions in a global oscillator
trap potential by adding a periodic potential with an incommensurate average ratio of number of ions to number of
periods being order of unity. With the increase of the periodic potential amplitude the system enters in the Aubry pinned
phase characterized by quasi-frozen positions of ions and a gap of their first phonon excitations becomes independent
of number of ions. This makes this quantum computer really scalable. It is augured that the usual single- and two-qubit
gates can be realized between the nearby ions in the Aubry phase. The possibilities of experimental realizations of a
periodic potential with microtrap arrays or optical lattices are discussed.

1 Introduction Here I propose another concept of quantum computer with

The creation of a scalable quantum computer for generic com-
putational tasks is an important challenge of modern quantum
technology [1]. One of the first physical proposals of such a
computer is the Cirac-Zoller quantum computer of 1995 with
a chain of cold ions placed in an oscillator trap potential [2].
Indeed, at that time the storage of cold ions already allowed
to keep several tens of ions in a storage ring [3]. Thus soon
after the proposal a two-qubit gate with a conditioned phase
shift had been realized [4] followed later by realization of a
few other two-qubit gates [5,6,7]. Simple quantum algorithms
[8], a set of universal gates with two ions [9] and a creation
of various entangled states [10] had been also reported. The
experimental progress with cold ion experiments is reviewed
in [11,12,13,14]. At present up to 100 ions can be routinely
trapped for hours in a linear trap configuration [15]. This exper-
imental progress makes cold ions to be very attractive for scal-
able quantum computer realization. Their important physical
advantages are related to possibilities of individual addressing
of a selected ion by a laser beam and low temperatures reached
experimentally.
However, the scalable quantum computation with ion-trap
computers is not so easy to reach even if about 100 ions can
be now trapped for hours. Thus, the original Cirac-Zoller pro-
posal [2] is not really scalable for a very large number of ions.
Indeed, the coupling between ion chain and the internal√ion lev-
els decreases with the number N of trapped ions as 1/ N (see
Eq.(1) in [2] ). Also, the ion chain oscillation frequency ωtr
is unavoidably decreasing if the number of ions in the trap is
growing with a constant average distance between ions. Thus
the gap between the ground state and the first excitation of ion
chain drops with N. It is proposed to avoid these problems
with a modular type architecture with quasi-separated groups
of ions with a further adiabatic transfer of quantum informa-
tion between groups. However, the practical realization of this
concept is not an easy task.
cold ions in a linear configuration based on the Aubry pinned
phase [16]. In this proposal the linear chain of ions is placed in
a periodic potential (or lattice), created by external fields, and a
global oscillator trap potential. It is assumed that there is an in-
commensurate density of ions ν = N/L ∼ 1.618 (ratio of num-
ber of ions N per number of potential periods L). In the limit
of small potential amplitude the system is reduced to the Cirac-
Zoller proposal. In this regime the spectrum of ion excitations
have an almost acoustic spectrum starting from ωtr which goes
to zero in the limit of large number of ions. However, when
the lattice amplitude K becomes larger than a certain critical
value Kc the chain enters in the Aubry pinned phase with the
appearance of optical gap ωg of excitations being independent
of the chain length and the number of ions placed in it. The
physics of this transition is related to the dynamical symplectic
maps, invariant Kolmogorov-Arnold-Moser (KAM) curves
and the fractal cantori replacing these curves above the tran-
sition to the Aubry pinned phase corresponding to the chaotic
map dynamics. Since the spectral gap ωg is independent of the
system size it is possible to place unlimited number of ions in
such a system.
The first analytical and numerical studies of ions in a peri-
odic potential had been done in [17] where its physical proper-
ties and the critical point of Aubry had been determined. The
cold ion experiments had been started in [18] and the signatures
of the predicted Aubry transition have been reported recently
by the Vuletic group [19,20] with up to 5 ions. The Aubry
phase with chains of larger number of ions is under investi-
gations [21]. Recently, the transport properties of charges in a
periodic 1D and 2D lattices have studied analytically and nu-
merically in [22,23]. However, in these studies [17,22,23] ions
or charges were considered without internal states while they
are essential since they form a qubit for a given ion and the
interactions between internal ion states (usually S and D states
are used that give a qubit lifetime of about a second [2,13]).
Also the coupling between internal ion states and spacial mo-
2 D.L. Shepelyansky: Quantum computer with cold ions in the Aubry phase

tion of ions is essential for the realization of universal quantum 1

gates. These features are discussed in this work with arguments
about the advantages of ions placed in a lattice of Aubry phase. 0.8
The paper is constructed as follows: the system description
and its physical properties are given in Section 2, the quantum
gates with ions in the Aubry phase are discussed in Section 3
0.6

ωk
and the discussion of the results and possible experimental re-
alizations are given in Section 4. 0.4

0.2
2 System description and properties
The motion of ions in a periodic potential and a global oscilla- 0
tor potential is described by the Hamiltonian [17]: 0 0.2 0.4 0.6 0.8 1
N
k
P2 ωtr 2 2 1
H = ∑( i + xi − K cos xi ) + ∑ . (1)
i=1 2 2 i> j i x j |
|x −

Here Pi , xi are ion momentum and position, K gives the am-

plitude of periodic potential and all N ions are placed in a
harmonic trap potential with frequency ωtr . The Hamiltonian
is written in dimensionless units where the potential period
is ℓ = 2π and ion mass and charge are m = e = 1. In these
atomic-type units the physical system parameters are expressed
in units: ra = ℓ/2π for length, εa = e2 /ra = 2π e2 /ℓpfor energy,
Eadc = εa /era for appliedp static electric field, va = εa /m for
particle velocity, ta = era m/εa for time t.
The physical properties of this system has been analyzed in
detail in [17]. They are not sensible to the boundary conditions
so that instead of global oscillator potential one can consider
the ion chin with fixed ends or hard wall boundary conditions
[23,24].
Fig. 1. (Color online) Phonon spectrum ω (k) as a function of
scaled mode number k = i/N (i = 0, . . . , N − 1) for the KAM
sliding phase at K = 0.03 (bottom curve, red squares) and the
Aubry pinned phase at K = 0.2 (top curve, blue points) for
N = 50 ions in a trap with frequency ωtr = 0.014 which ap-
proximately gives the golden mean density in the central 1/3
part of the chain (after [17]).
0.4
0.3
ω0(K)

The equilibrium positions of ions are determined by the -1

condition Pi = 0 and ∂ H/∂ xi = 0. In the approximation of in- 0.2 10
teractions only between nearest neighbors this give the recur- -2
sive map for equilibrium ion positions xi : 10
0.1
√ 10
-3
pi+1 = pi + Kg(xi) , xi+1 = xi + 1/ pi+1 . (2)
0 0.035 0.07
0
)2
Here pi = 1/(xi − xi−1 is the effective momentum conjugated 0 0.1 0.2 0.3
to xi and the kick function is Kg(x) = −ω 2 x − K sin x. The nu-
merical simulations performed in [17,22,23,24] confirm that K
this approximation provides a good description of real ion po-
sitions obtained by numerical simulations. Thus the nearest
neighbor interactions between ions are dominant.
The map description (2) provides important links with the Fig. 2. (Color online) Minimal excitation frequency ω0 (K)
generic properties of dynamical symplectic maps (see e.g. [25,26,27]).
as a function of periodic potential strength K for the golden
The equation for xi+1 can be locally linearized in pi+1 near the mean ion density νg = 1.618... and number of ions N = 50
resonant values of pr ≈ 2π /ν defined by the condition xi+1 = (red squares; ωtr = 0.014), N = 150 (magenta circles; ωtr =
xi + 2π m where m are integers (see examples in [25,26]). This 0.00528), N = 300 (blue triangles, ωtr = 0.00281). The critical
leads to the local description of dynamics by the Chirikov stadard point Kc ≈ 0.05 of Aubry transition is marked by arrow; inset
map [17]: shows data near Kc (after [17]).
yi+1 = yi − Ke f f sin xi , xi+1 = xi − yi+1 , (3)

where yi = α (pi − pr ), α = 1/(2pr 3/2 ) = (2π /ν )3 /2 and the system of ions in a periodic potential can be understood from
dimensionless chaos parameter Ke f f = α K = K(2π /ν )3 /2. the properties of the Chirikov standard map which describes
This local description corresponds to the linear-spring forces the local dynamics of various physical systems (see e.g. [29]).
locally acting between particles that in fact represents the Frenkel- At small K or Ke f f the phase space of maps (2) and (3)
Kontorova model describing commensurate-incommensurate tran-is covered by the invariant KAM curves characterized by irra-
sition in solid states systems [28]. Thus the properties of this tional rotation number r =< xi − x0 > /2π i = ν which gives
 D.L. Shepelyansky: Quantum computer with cold ions in the Aubry phase
an average distance (phase) between ions related to the aver-
age ion density ν . The oscillations of ions near the equilibrium
positions has the acoustic excitation spectrum ωk ≈ Cv k + ωtr
where k = i/N plays the role of wavevector number and Cv ∼ 1
is the sound velocity.
For the Chirikov standard map the last invariant curve with
the golden mean rotation number rg = νg = 1.618... is destroyed
at Ke f f ≈ 1 leading to a global chaos and diffusion in y [25]. For
the case of ions with density ν this gives the critical amplitude
of potential [17]:
Kc (ν ) ≈ 0.034(ν /νg)3 , νg = 1.618... . (4)
This theoretical dependence is recently confirmed by extensive
numerical simulations [23]. For ν = νg the numerical results
give Kc = 0.0462 [17,23,24] that is slightly higher than the the-
oretical value due to the local approximation used in the reduc-
tion to the Chirikov standard map.
For K > Kc (ν ) the invariant KAM curve is destroyed and it
is replaced by a fractal cantori invariant set as proved by Aubry
in [16]. The configuration of particles corresponding to this
invariant set has the minimal energy and thus represents the
ground state of the system. The spectrum of ion oscillations
near these√ ground state positions is characterized by the optical
gap ωg ∼ K. Thus in difference from the KAM sliding phase
at K < Kc for K > Kc we have the Aubry pinned phase where
the ion chain is pinned by the lattice.
The example of excitation spectrum for the KAM and Aubry
phases is shown in Fig. 1 taken from [17]. The dependence of
the minimal excitation frequency ω0 (K) on potential amplitude
K is shown in Fig. 2 taken from [17]. For these data the trap fre-
quency ωtr is chosen in such a way that, at a given number of
ions N in the trap, the central 1/3 part of the chain keeps the
fixed density ν ≈ 1.618 when the number of ions N is growing.
Due to this condition at K = 0, corresponding to the Cirac-
Zoller proposal
√ [2], the trap becomes more and more soft and
ωtr ∼ 1/ N → 0. Indeed, we want to keep the distance be-
tween ions in the center to be independent of N and thus size of
the chain xchain ∼ N/ν is growing since it is approximately de-
2
termined by the condition at the chain end Fchain
2
√ ∼ ωtr xchain ∼
ν that gives the above dependence ωtr ∼ 1/ N.
Thus for K < Kc the lowest excitation frequency goes to
zero with the increase of number of ions in the trap. Hence
the Cirac-Zoller proposal in not really scalable. In contrast for
K > Kc the lowest frequency excitation in independent of N as
it is well seen in Fig. 2. Thus this Aubry pinned phase has cer-
tain chances to represent a scalable architecture for a quantum
computer with cold ions.
Indeed, for the quantum case the energy of lowest phonon
excitation is E0 = h̄ω0 (K) = h̄ωg being independent of N. For a
temperature T ≪ h̄ω0 (K) the phonon excitations become frozen
and should not perturb the accuracy of quantum gates opera-
tions.
There are also another type of quantum excitations in the
quantum ion chain inside the Aubry pinned phase. In fact the
Aubry theorem [16], which guaranties that the Aubry cantori
ground state has the minimal energy EA of the classical ion
chain is mathematically correct but it is wrong from the phys-
ical view point. Indeed, in the classical chain there are ex-
ponentially many static configurations of ions which number
3
Ns grows exponentially with the number of trapped ions N.
In addition the energies of these configurations are approach-
ing exponentially close to the Aubry ground state energy EA
with increase of N (see Fig. 4 in [17] where this feature is
clearly illustrated). In fact this property is similar to the ran-
dom spin glass systems [30]. However, in our case the random-
ness is absent and the system is described by a rather simple
deterministic Hamiltonian (1). Thus the Aubry pinned phase
represents the dynamical spin glass system with an enormous
amount of quasi-degenerate configurations in a vicinity of the
Aubry ground state.
In the quantum case there is quantum tunneling between
these quasi-degenerate configurations that can be viewed as an
instanton excitations. However, for small dimensionless Planck
constant h̄e f f the gas of instantons is very dilute and the tunnel-
ing times are enormously long [17]. Thus on a scale of typical
tunneling time ttul ∝ exp(A/h̄e f f ) we can consider the ions to
be frozen at their positions (here A ∝ K is a typical action be-
tween energy minima coupled by tunneling). The dimension-
p
less Planck constant is h̄eff = h̄/ (e mℓ/2π ) and for a typical
lattice period ℓ ≈ 1µ m, ion density ν ∼ 1 and 40 Ca+ ions we
have very small h̄eff ≈ 10−5 . Thus the quantum ions can be con-
sidered as frozen at their configuration positions for the whole
time scale of quantum computations.
3 Quantum gates
As in the proposal of Cirac-Zoller [2] I assume that the qubit is
formed by two internal levels S1/2 and D5/2 of 40 Ca+ ion with
a radiative life-time of more than one second. All single-qubit
gates can be realized by laser pulses as described in [2,13].
At present these gates are routinely performed with the fidelity
exceeding 0.99 [13]. The individual accessing of ions is also
available in experiments with ion spacing of about 5 µ m [13].
Since single-qubit gates with ions are reliable the most im-
portant for quantum computations become two-qubit gates which
in combination with single-qubit gates allow to perform univer-
sal quantum computations [1]. There are three types of two-
qubit gates usually discussed for cold ions (see e.g. review
[13]): the Cirac-Zoller gate [2], the Molmer-Sorensen gate [31]
and the geometric phase gate [7] closely related to the Molmer-
Sorensen gate.
In all these gates the motional oscillator states of ions (side-
band) with frequency ω0 are coupled by a tuned laser pulse
with internal S − D levels of ions. Usually as an example one
considers two ions with two internal levels and their sideband
modes [13]. The laser pulse duration is selected in a way allow-
ing to perform two-qubit gate. In the case of long ion chain in
an oscillator trap the operational frequency of the Cirac-Zoller
gate is proportional to the strength of coupling of internal lev-
els with the whole chain oscillator state√ (the bus mode) which
decreases with the number of ions as 1/ N.
For the Molmer-Sorensen gate both ions are irradiated with
a bichromatic laser field with frequencies ω0 ± (ωqubit + δ )
tuned close to the red and the blue sideband of a collective
mode (see Fig.14 in [13]). This approach allowed to create ex-
perimentally Bell states with a fidelity 99.3% [13]
The same gates can be implemented for ions in the Aubry
pinned phase. In this case the interaction of ions is well approx-
4 D.L. Shepelyansky: Quantum computer with cold ions in the Aubry phase
imated by the nearest neighbor interactions as is discussed in
the previous Section with the map (2) description of equilib-
rium ion configurations. The oscillations of ions in a vicinity
of equilibrium positions is harmonic and we can consider them
as sideband transitions for laser pulses as for the two-qubit
gates considered above. Since the interactions are dominated
by nearest neighbors the coupling between internal S − D qubit
levels and ion oscillator mode is independent of the number of
ions in the chain. The frequency of this oscillator or phonon
mode gap is ωg = ω0 (K) being also independent of the chain
length as it is shown in Fig. 2.
The construction of two-qubit gates should also take into
account that when cold ions are cooled and loaded in the Aubry
pinned phase it is most probable that they will be located in one
of quasi-degenerate static configurations. Thus the distances
between nearby ions will be somehow irregular that will af-
fect the interactions between specific pairs of ions. However, it
is possible to determine experimentally the actual ion positions
and then to adapt the laser pulses of two-qubit gates to these
experimentally found ion positions. In a sense for a good work
of a piano each string should be checked and adapted. Here, for
quantum gates with ions in the Aubry phase we have a similar
situation.
However, there are certain points that require additional in-
vestigations. The low energy phonon exciations with the low-
est phonon frequency ω0 (K) are excited by a tuned laser pulse
which acts mainly on one or two nearby ions. Thus there is a
question how this excitation will propagate along the chain of
ions in the Aubry pinned phase. This propagation or spread-
ing along the chain depends on two main factors: the local-
ization properties of phonon modes in the pinned phase and the
rate of decomposition of local ion oscillations into these photon
modes. At present very little analysis has been performed for
these important properties of ionic phonon modes in the Aubry
phase. Examples of a few phonon eigenmodes are given in [17]
(see Figs.9,10 there). Some of modes look to be localized some
of them have spreading over several ions. The spreading rate of
one or two ion oscillations has not been studied and require
further investigations.
Thus there are open questions on the possible fidelity and
accuracy of two-qubit gates for cold ions in the Aubry phase.
4 Discussion
In this work I analyzed the properties of cold ion chain in a pe-
riodic potential which amplitude locates ions inside the Aubry
pinned phase. The emergence of Aubry transition from KAM
sliding phase to Aubry pinned phase takes place when the po-
tential exceeds a critical value VA = Kc (ν )e2 /(ℓ/2π ). For a typ-
ical lattice period ℓ = 1 µ m and dimensionless ion density per
period ν = 1.618 this corresponds to VA ≈ 3kB Kelvin. Appar-
ently this amplitude significantly exceeds the amplitudes reach-
able with presently available laser power for optical lattices.
Usually the optical lattice amplitude is assumed to be able to
reach the value VA ≈ 10−3kB Kelvin (see e.g. review [32]).
In contrast to optical lattices the radio-frequency (RF) traps
provide the potential depth VRF ≈ 104 kB Kelvin that is signifi-
cantly above the estimated Aubry transition potential amplitude
[32]. At present there is a significant miniaturization of these
RF traps with sizes going down to tens of microns [32]. Thus
such microtrap linear arrays can model the periodic potential
considered here with high amplitudes of periodic potential al-
lowing to place ions in the Aubry pinned phase. There is also
progress with the Penning mircotraps of about 10 micron size
[33]. Thus the linear array of RF or Pinning microtraps would
allow to observe the Aubry transition and hopefully to perform
scalable quantum computations with cold ions in the Aubry
pinned phase.
5 Acknowledgments
I thank P.Zoller for discussion and pointing to me Refs. [32,33].
This work was supported in part by the Pogramme Investisse-
ments d’Avenir ANR-11-IDEX-0002-02, reference ANR-10-
LABX-0037-NEXT (project THETRACOM).
References
1. M.A. Nielsen and I. Chuang, Quantum computation and quantum
information, Cambridge University Press, Cambridge UK (2000).
2. J.I. Cirac and P. Zoller, Quantum computations with cold trapped
ions, Phys. Rev. Lett. 74, 4091 (1995).
3. G. Birkl, S. Kassner, and H. Walther, Multiple-shell structures of
laser-cooled 24 Mg+ ions in a quandrupole storage ring, Nature
357, 310 (1992).
4. C. Monroe, D.M. Meekhof, B.E. King, W.N. Itano, and
D.J. Wineland, Demonstration of a fundamental quantum logic
gate, Phys. Rev. Lett. 75, 4714 (1995).
5. C.A. Sackett, D. Kielpinski, B.E. King, C. Langer, V. Meyer,
C.J. Myatt, M. Rowe, Q.A. Turchette, W.M. Itano, D.J. Wineland,
and C. Monroe, Experimental entanglement of four particles, Na-
ture 404, 256 (2000).
6. B. DeMarco, A. Ben-Kish, D. Leibfried, V. Meyer, M. Rowe,
B.M. Jelenkovic, W.M. Itano, J. Britton, C. Langer, T. Rosenband,
and D.J. Wineland, Experimental demonstration of a controlled-
NOT wave-packet gate, Phys. Rev. Lett. 89, 267901 (2002).
7. D. Leibfried, B. DeMarco, V. Meyer, D. Lucas, M. Barrett, J. Brit-
ton, W.M. Itano, B. Jelenkovic, C. Langer, T. Rosenband, and
D.J. Wineland, Experimental demonstration of a robust, high-
fidelity geometric two ion-qubit phase gate, Nature 422, 412
(2003).
8. S. Gulde, H. Haffner, M. Riebe, G. Lancaster, C. Becher, J. Es-
chner, F. Schmidt-Kaler, I.L. Chuang, and R. Blatt, Quantum in-
formation processing with trapped Ca+ ions, Proc. R. Soc. Lond.
A 361, 1363 (2003).
9. F. Schmidt-Kaler, H. Haffner, M. Riebe, S. Gulde, G.P.T. Lan-
caster, T. Deuschle, C. Becher, C.F. Roos, J. Eschner, and R. Blatt,
Realization of the CiracZoller controlled-NOT quantum gate, Na-
ture 422, 408 (2003).
10. C.F. Roos, M. Riebe, H. Haffner, W. Hansel, J. Benhelm,
G.P.T. Lancaster, C. Becher, F. Schmidt-Kaler, and R. Blatt, Con-
trol and measurement of three-qubit entangled states Science 304,
1478 (2004).
11. R. Blatt and D. Wineland, Entangled states of trapped atomic
ions, Nature 453, 1008 (2008).
12. M. Johanning, A.F. Varon and C. Wunderlich, Quantum simula-
tions with cold trapped ions, J, Phys. B: At. Mol. Opt. Phys. 42,
154009 (2009).
 D.L. Shepelyansky: Quantum computer with cold ions in the Aubry phase 5

13. R. Blatt and C.F. Roos, Quantum simulations with trapped ions,
Nature Phys. 8, 277 (2012).
14. D.J. Wineland, Nobel Lecture: Superposition, entanglement, and
raising Schrodinger’s cat, Rev. Mod. Phys. 85, 1103 (2013).
15. G. Pagano, P.W. Hess, H.B. Kaplan, W.L. Tan, P. Richerme,
P. Becker, A. Kyprianidis, J. Zhang, E. Birckelbaw, M.R. Her-
nandez, Y. Wu and C. Monroe, Cryogenic trapped-ion system
for large scale quantum simulation, arXiv:1802.03118 [quant-ph]
(2018).
16. S. Aubry, The twist map, the extended Frenkel-Kontorova model
and the devil’s staircase, Physica D 7 (1983) 240.
17. I. Garcia-Mata, O.V. Zhirov, and D.L. Shepelyansky, Frenkel-
Kontorova model with cold trapped ions, Eur. Phys. J. D 41, 325
(2007).
18. T. Pruttivarasin, M. Ramm, I. Talukdar, A. Kreuter, and
H. Haffner, Trapped ions in optical lattices for probing oscilla-
tor chain models, New J. Phys. 13, 075012 (2011).
19. A. Bylinskii, D. Gangloff, and V. Vuletic, Tuning friction atom-
by-atom in an ion-crystal simulator, Science 348, 1115 (2015).
20. A. Bylinskii, D. Gangloff, I. Countis, and V. Vuletic, Observation
of Aubry-type transition in finite atom chains via friction, Nature
Mat. 11, 717 (2016).
21. J.Kiethe, R. Nigmatullin, D. Kalincev, T. Schmirander, and
T.E. Mehlstaubler, Probing nanofriction and Aubry-type signa-
tures in a finite self-organized system, Nature Comm. 8 15364
(2017).
22. M.Y. Zakharov, D. Demidov and D.L. Shepelyansky, Wigner
crystal diode, arXiv:1901.05231 [cond-mat.mes-hall] (2019).
23. O.V. Zhirov, J. Lages, and D.L. Shepelyansky, Thermoelec-
tricity of cold ions in optical lattices, arXiv:1901.09588[cond-
mat.quant-gas] (2019).
24. O.V. Zhirov, and D.L. Shepelyansky, Thermoelectricity of Wigner
crystal in a periodic potential, Europhys. Lett. 103, 68008 (2013).
25. B. V. Chirikov, A universal instability of many-dimensional oscil-
lator systems , Phys. Rep. 52 (1979) 263.
26. A.J.Lichtenberg, M.A.Lieberman, Regular and chaotic dynamics,
Springer, Berlin (1992).
27. J.D. Meiss, Symplectic maps, variational principles, and trans-
port, Rev. Mod. Phys. 64(3), 795 (1992).
28. O.M. Braun and Yu.S. Kivshar, The Frenkel-Kontorova Model:
Concepts, Methods, Applications, Springer-Verlag, Berlin (2004).
29. B. Chirikov and D. Shepelyansky, Chirikov standard map, Schol-
arpedia 3(3), 3550 (2008).
30. N. Mezard, G. Parisim and M.A. Virasoro, Spin glass theory and
beyond, World Sci., Singapore (1997).
31. A. Sorensen, and K. Molmer, Quantum computation with ions in
thermal motion, Phys. Rev. Lett. 82, 1971 (1999).
32. Ch. Schneider, D. Porras, and T. Schaetz, Experimental quantum
simulations of many-body physics with trapped ions, Rep. Prog.
Phys. 75, 024401 (2012).
33. S. Jain, I. Alonso, M. Grau, and J.P. Home, Quantum
simulation with ions in micro-fabricated Penning traps,
arXiv:1812.06755[quant-ph] (2018).